Explorar skills
43 skills encontradas
work-report-generator
Generate formal daily work reports from unstructured input. Asks output language once, processes one date per invocation, never infers tasks or hours, requires explicit close confirmation before generating, keeps `reports/projects.md` + `reports/YYYY-MM-DD/{LOG,REPORT}.md`. Use on "reporte de trabajo", "daily report", or /work-report.
grpc-architect
Vanilla gRPC standards — .proto services, status.Error with standard codes, domain→code mapping, interceptor chain (auth/log/recovery/validation/metrics), client deadlines, context propagation, reflection off in prod, bufconn testing. Language-agnostic; Go examples. Use when designing or reviewing a gRPC service.
mcp-architect
MCP (Model Context Protocol) 2025-11-25 server standards — tool/resource/prompt primitives, capability negotiation, Streamable HTTP transport with Mcp-Session-Id, OAuth 2.1 + RFC 8707 resource indicators, tool annotations (readOnly/destructive/idempotent), structured output, JSON-RPC error mapping, prompt-injection and SSRF defenses, MCP Inspector testing. Python (FastMCP) and Go (official SDK) re
rest-api-architect
Cross-language REST conventions — resource URLs, method semantics, status codes, URL-prefix versioning, cursor pagination, snake_case JSON, ISO 8601 timestamps, RFC 7807 errors, Idempotency-Key, ETag/If-Match, OpenAPI as source of truth. Framework-agnostic. Use when designing or auditing REST endpoints.
api-contract-reviewer
Reviews REST + gRPC contracts for stability, versioning, completeness, backwards compatibility. References rest-api-architect / protobuf-architect / grpc-architect for rules; runs `buf breaking` / `openapi-diff`. Severity-keyed findings. Use when reviewing a new endpoint, proto change, or before a breaking-change release.
performance-reviewer
Cross-language perf review — N+1, missing indexes, blocking I/O in async, allocation hot paths, unbounded memory, slow algorithms. Findings grounded in EXPLAIN / pprof / py-spy / metrics. Use when reviewing for perf or investigating a slow endpoint.
security-reviewer
Cross-language security review — injection, auth/authz, secrets, insecure defaults, deserialization, CSRF/SSRF/IDOR, dep vulns. Emits a Critical/High/Medium/Low report with file:line + fixes. Use when auditing a PR or pre-release.
code-design-refactor
Design-level refactoring rules — extraction, decoupling, SRP, encapsulation, primitive obsession. Language-agnostic. Sits between logic-cleaner (expression-level) and improve-codebase-architecture (system-level). Use when restructuring existing code at the module/class/function level.
design-patterns
Skeptical design-patterns reference catalog for modern Go and Python — keeps repository, adapter, strategy, decorator, observer, builder; flags the rest as anti-patterns. Consultative, not active: consult before introducing a pattern in code-design-refactor or improve-codebase-architecture, or when pushing back on a cargo-cult one. Not a refactoring rung itself.
improve-codebase-architecture
Find deepening opportunities in a codebase, informed by the domain language in CONTEXT.md and the decisions in docs/adr/. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase more testable and AI-navigable.
logic-cleaner
Expression-level code cleanup — guard clauses, naming, magic values, conditional simplification, duplication. Language-agnostic. Out of scope: architecture, module decomposition. Use when user asks to clean code, simplify expressions, flatten nesting, or invokes /logic-cleaner.
ros2-architect
ROS2 standards across Jazzy/Kilted/Lyrical — colcon + ament_cmake/ament_python layout, lifecycle nodes, explicit QoS, services/actions, parameters, Python launch DSL. Cross-platform dev env via Pixi + RoboStack. C++ (rclcpp) and Python (rclpy) equal first-class. Use when scaffolding or reviewing a ROS2 workspace.